FLOWS OF VOLCANIC MUD. 279 



lets along this line westward are the narrow canyons cut by the present 

 ] Jear, Steep Hollow, G reenhorn and South Yuba rivers. Through none of 

 these canyons is there the remotest possibility that the ancient river passed 

 westward. High bed-rock is exposed in each case on both sides of 

 the gap cut by the recent rivers, and not even the smallest remains of 

 any Neocene deposits are met with for some distance below any of the 

 supposed gaps. Any such supposition would, moreover, necessitate ex- 

 tremely improbable and curious bifurcations. Along the whole line, 

 Dutch Flat to Badger hill, there are gravels accumulated to an excep- 

 tional depth and extent. Above these gravels rest, at many places, the 

 remnants of an eroded Mow of rhyolitic tuffs and sands. They are firs! 

 met with in Canyon creek, about five miles above Alta. They are ex- 

 posed at Shady run, at Alta, at You Bet, at Hunts hill, at Buckeye bill 

 and at Quaker hill. Again, they are exposed at Seotts Hat, on the north- 

 ern side of Deer creek, and, finally, at Blue Tent, where their volcanic 

 character begins to be less apparent, being largely mixed with other 

 detrital material; but everywhere they form a sheet perhaps a, hundred 

 feet thick and resting on several hundred feet of gravel. 



The continuity and the direction of these Hows of rhyolitic mud are 

 distinctly and unmistakably indicated on the geologic map. Coming 

 down the old channel along the upper course of Canyon creek, they 

 flowed in a southwesterly direction, passing Alta, and Dutch Flat, down 

 to You Bet. Turning here with the valley, they flowed northward by 

 Quaker hill to Blue Tent. Between Blue Tent and Badger hill the vol- 

 canic masses are completely eroded and the underlying gravel beds 

 exposed. 



Mr. Pettee traced the deep channel northward as far as Hunts hill, or 

 even, with a somewhat uncertain elevation, to Quaker hill. If he had 

 examined the relations at Seotts Hat and the country between Seotts 

 flat and Blue Tent, I am confident he would have arrived at the same 

 conclusion which has been reached here. At Seotts Hat there is the 

 most ample evidence that a very large channel crosses Deer creek, with 

 rapidly rising rim-rock on the east and west. The creek has not quite 

 <ait through the ancient river bed, the bed-rock being covered for a dis- 

 tance of nearly a quarter of a mile. The same accumulations of sands 

 and tuffs as are exposed at Quaker hill arc found abundantly above 

 Seotts Hat and in Rock creek between Seotts flat and Blue Tent. It 

 has been shown that the deep channel as far as Hunts hill has no pos- 

 sible westerly or southerly outlet. High bed-rock all along on the west- 

 ern as well as the eastern side from here northward simply makes any 

 other outlet than by Blue Tent impossible, if we do not assume entirely 

 improbable faults of several hundred feet of throw. There is no point 



